Top U.S.-based vehicle manufacturers learned to shift more of their warranty expenses back onto their suppliers during the recession, but then massive recalls created additional expenses they haven't yet recovered. In looking at the pattern of warranty expense data since 2003, the relative shares of the OEMs and their suppliers looks a bit like a valley between two mountains.

This week, we're looking at the success rate of automotive supplier recovery efforts, as measured by the relative distribution of warranty expenses between the OEMs and their suppliers. To perform this analysis, first, we selected a list of 55 automotive OEMs and 122 automotive parts and component suppliers that were based in the U.S. and were reporting their warranty expenses in their financial statements. Some were making passenger cars and motorcycles, while others were making trucks and heavy equipment. Then we collected three essential metrics from each: the amount of claims paid, the amount of accruals made, and the amount of warranty reserves held at the end of each quarter since 2003.

The timing is almost perfect for collecting all the warranty data for the first half of 2016, since most companies have filed their financial statements in the past few weeks. So we have 54 quarters of data to work with. And it wasn't until a reader inquired that we realized it's been almost exactly two years since we last ran a supplier recovery report (Warranty Week, August 21, 2014). So the topic is overdue for an update.

What we did back then and what we'll do again this week is look at the shifting burden between the OEMs and the suppliers when it comes to warranty expenses, based not only on dollars but also on percentages of the total. But rather than look at just claims totals, we're also going to look at accruals and reserves.

By now, all of the first-half-2016 data has been collected, with the exceptions of Briggs & Stratton Corp. and Twin Disc Inc., which are still a week or two away from filing their annual reports (which usually arrive a little later than the quarterly financial statements). So we have only 53 quarters of data to work with for those companies. And we're completely excluding Chrysler and FCA US, because the company's warranty expenses are rolled up into its European parent company's totals, and are no longer reported separately.

Net Warranty Expense Reports

The theory is that since all these companies are reporting their net expenses (payouts minus reimbursements), the totals for the entire industry must represent 100% of the U.S. warranty expense. That's not entirely true, because many U.S. firms export, and also because international automotive manufacturers such as Toyota, Honda, VW, Hyundai, and Autoliv produce within the U.S. But the model is complex enough without complicating it further with cross-border adjustments.

In the short version of the story, we've found that over the past 14 years, the customer-facing OEMs paid about 85% of the industry's total warranty expenses, while their suppliers paid the remaining 15%. But in the long version, the OEMs paid a decreasing share that fell from roughly 90% in 2003 to somewhere around 80% by the time the great recession hit. Their share has been slowly rising again ever since, peaking again in 2014.

In Figure 1, we've provided the totals for claims paid by the OEMs and the suppliers over the past 54 quarters. There are two peaks: $3.61 billion in the last quarter of 2007, and $3.53 billion in the middle of 2014. And the trough, of $2.44 billion, came in the second quarter of 2010.

In that fourth quarter of 2007, the OEMs paid $3.16 billion while the suppliers paid $451 million, so their relative shares of the total were 87% and 13%, respectively. In Figure 2 we've made similar calculations for all the other quarters, using the data from Figure 1, and we've zoomed in on just the range that contains the shifting line of division between the two segments, which in this case is 79% to 91%.

It turns out that late 2007 represented one of the last instances where the OEMs paid that much of the total. The all-time peak was just under 90% in the middle of 2004. The OEMs' share remained in a range of 87% to 90% from 2003 to mid-2008, before it began to fall. It hit a low of 81% in 2012 before it began to rise again. And then it peaked again at 89% in 2014 before slipping back in the quarters since.

It was in the years after the recession struck that we began to hear stories coming out of Detroit that the OEMs were very consciously and deliberately trying to increase the success rate of their supplier recovery efforts. That explains the 2008-to-2014 shift quite well. And then in 2014, a series of massive product recalls hit the OEMs, and they also lightened up on their supplier recovery efforts, so their share began to grow again.

Interestingly, in 2012, at the moment the OEMs shifted the most warranty expense to their suppliers, those suppliers were paying the most warranty expense. In only four of the 54 quarters did the suppliers pay out more than a net $500 million in claims, and that was their peak ($519 million). Two of the other $500+ million quarters were in 2011, so by the looks of it, that was the peak period for supplier recovery success, at least in terms of claims paid.

Warranty Accrual Shares

But we have two other warranty metrics to examine. In Figure 3 we're counting up all the accruals made by the 55 OEMs and 122 suppliers in the 54 quarters since the beginning of 2003. The pattern is a little different, with the peaks coming in both 2005 and 2007 and then the trough coming in 2009. Accruals fell by half, from a $3.7 billion quarterly peak in mid-2005 to a $1.8 billion quarterly trough in early 2009.

A period of gradual recovery followed. But then the cost of the recalls hit in 2014, particularly the General Motors ignition switch defect, driving industry accruals to a new peak of $3.85 billion. It's fallen back to the 2010-2013 range in the quarters since.

Compared to Figure 1, both the peaks and troughs in Figure 3 are more dramatic. That's probably because accruals are something a company decides to do, based on sales projections and expectations of repair costs, while claims are something that happens to them, based on customer usage and other factors.

More than anything, accruals are made immediately at the time of sale. So in years like 2009 when sales are few, the accrual totals also plummet. In contrast, with vehicles whose warranties last multiple years, the claims coming in from recent sales mix with claims from year-ago and two-year-ago sales. So there's more of an evening out of the peaks and valleys.

We say this because in Figure 4, which turns the dollar totals of Figure 3 into percentage shares for the OEMs and suppliers, the peaks and valleys for the two groups are even more dramatic. The first peak of 2004 is 91%, followed by the trough of 2009 at 77%, then the second peak comes in 2014 at 89%.

It can't be blamed on the suppliers. In fact, the suppliers' accrual total was last above $500 million or below $400 million in 2010. In contrast, the OEMs' accrual total has swung widely, impacted by recessions and recalls. So what we're seeing here with this metric is a relatively steady total for the suppliers contrasted against a volatile and quickly reactive total for the OEMs.

Warranty Reserve Balances

Next, we'll look at the ending balances in the warranty reserve funds of the 55 OEMs and 122 suppliers. And it's a balance that is the product of the claims paid and the accruals made, along with other adjustments for acquisitions, foreign exchange fluctuations, and corrections for past over- or under-accruals that were detected as time went by. So it's no surprise that it changes more slowly than either claims or accrual totals.

In Figure 5, the industry's initial peak reserve balance comes in 2007 at $22 billion, and the trough is in 2010 at $16.3 billion. And then, as the recovery progresses, the balance again grows, peaking at $24.2 billion during the height of the recalls, before falling back again.

For suppliers, the balance grew gradually from 2003 to 2008, peaking at $3.1 billion in early 2008 before falling as low as $2.8 billion in mid-2009. But since 2010 it's remained mostly in a range of $3.3 to $3.4 billion, with little change from one quarter to the next.

In contrast, the OEMs' reserves have gone up or down dramatically. For instance, their collective balances fell by nearly $6 billion from the end of 2007 to the end of 2010. Then they rose by $7.25 billion over the next four years. So again, what we're seeing is a relatively steady balance for the suppliers and a highly volatile balance for the OEMs.

This shows up clearly in the percentage shares in Figure 6, with some minor differences. For instance, the greatest share for the OEMs comes in early 2003, at just under 89%, and over the following eight years it falls all the way down to just under 80%. In 2014, it briefly surpasses 86% before falling back again.

Therefore, over 54 quarters, the OEMs' share of claims has ranged from 81% to 90% of the industry total. Their share of accruals has ranged from 79% to 91%. And their share of reserves has ranged from 80% to 89%. Meanwhile, of the 162 ratios, only five times have the suppliers' share been above 20% and only three times has it been below 10%. In other words, a total of more than 19 out of 20 of these measurements have been between those two extremes: 80/20 and 90/10.

The 80/20 Rule?

This suggests that there's some truth to the perspective we've gotten from some of the auto industry's leading experts on supplier recovery: the best they can ever hope for is to recover 20%, more or less. Based on the value of the parts compared to the value of the whole, and based upon the way that parts and labor costs are allocated between OEMs and suppliers, the best they can hope for is to shift 20% of the total warranty burden onto the backs of their suppliers.

It may be higher than 20% for truck and bus OEMs, since the suppliers of major components such as diesel engines are likely to pay dealerships directly for the labor portion of any warranty work. It may be lower for some types of vocational vehicles and heavy equipment, because suppliers provide a smaller share of the vehicle. But it's probably close to 20% for passenger cars, SUVs, vans and light trucks.

Secondarily, from the looks of the data above, and how accruals are usually made before claims are paid, it seems like the totals and balances of the OEMs react first, and react the most, to disruptions such as recessions and recalls. When a recession hits, they cut their warranty accruals the most and the fastest, while the suppliers change the least and the slowest. And these recessionary periods are the instances when the ratio approaches 80/20, as it did in 2009 and 2010.

At the other extreme, when a massive recall hits, the OEMs react first, and react the most, by raising their warranty accruals the most and the fastest, while the suppliers change the least and the slowest. And these are the instances when the ratio approaches 90/10, as it did in 2004 and 2014.

Of course, all of this is based on the perceptions of an external observer. Does this match the actual experience of the warranty professionals inside the companies involved? We've had some confidential discussions with a handful of supplier recovery experts who suggest that the 20% level is the peak. But there could be exceptions, and there could be other reasons for the shifting shares. Readers who wish to discuss this issue are welcome to confidentially contact the editor by phone or email.

Vacation Break

Meanwhile, it's that time of year again for end-of-summer vacations. This year, Warranty Week will shut down for one week, with no newsletter scheduled for Thursday, September 1, 2016. We will reopen on September 6, with the next newsletter going out on September 8. Until then, wishing all of you a cool breeze, chilled drinks, and a gentle current.